2019 CSD Releases

This page shows archive information on the 2019 releases.



For the most current release check the What's New page.

2020.0 CSD Release (December 2019)

Data Updates:

In the last year we have seen the released CSD grow to over one million structures for the first time; a huge milestone in structural chemistry. The continued growth in data and the increasing diversity and complexity of new structures means that users can have more confidence in their results and gain even more insights from the database.

This year’s release:

  • Takes the CSD to 1,016,168 unique structures (1,034,174 entries)
  • Comprises over 60,000 entries new entries
  • Includes over 5,000 new CSD Communications
  • Targeted enhancements to existing CSD entries
  • Read more about this data release



  • Further optimisation of our GOLD protein ligand docking software for use in ultra-large docking projects including libraries of tens of millions of compounds, enabling the running of the GOLD docking algorithm using fast settings on cloud deployed computing systems, with automated selection of the best results based on user-defined requirements.


  • Quick and easy hydrogen-bond likelihood analysis with new H-bond Coordination Quick-View, enabling the quick assessment of the likelihood of H-bond behaviour based on H-bond coordination numbers in the observed structure. Read more.


  • The incorporation of Python 3 by default, meaning the CSD Python API is Python 3 enabled straight away at the point of installation and easy to integrate with other key scientific Python packages such as TensorFlowscikit-learnmatplotlibpandas and RDKitRead more.

Product Telemetry:

  • The implementation of an entirely opt-in Product Telemetry system in our software which is able to capture information about which components have been used and on what kind of operating system, helping us to make better data-driven decisions to inform the direction of our software and ensure maximum value for you as a user. Find out more.

Licensing updates:

  • The launch of a new licensing system which increases the flexibility of access to CSD data and software. The new system is a modern, secure and regularly updated third-party licensing system which enables both online and offline activation, allows use of any software release with a valid licence, and provides flexibility in length of licence. Read more.

2019.3 CSD Release (September 2019)

Data Updates:

  • 14,673 new entries
  • This release brought the total size of the CSD to 1,023,815 entries and 1,006,180 unique structures, taking the CSD past one million structures for the first time!
  • Read more about this data release.


  • The launch of a new scripting interface that connects CCDC’s first class 3D substructure searching with searching of protein-ligand binding sites derived from the Protein Data Bank (PDB). This update enabled seamless interchange between searching CSD data and PDB data, allowing for the easy extraction of data driven insight from mining of 3D information including ligands with specific structures/similar structures, patterns of interactions between protein binding sites, waters, metals and small molecules, and geometric patterns within small molecules bound to proteins.
  • This release also included stage one development of GOLD software to deliver a streamlined high throughput mode for ultra-large docking, enabling reduced output which will improve speed and disk space across many platforms.

CSD-Materials & CSD-Discovery:

  • The launch of molecular interaction maps in the CSD Python API enabling automated generation of interaction maps for protein binding sites, automated generation of interaction maps for small molecules, and easy integration of interaction maps with 3rd party tools.


  • Polymer expansion in Mercury enabling much faster and more responsive expansion of polymeric, or networked, structures. Offers new options for more tailored expansion to explore structures the way you want to and easy pruning of structural views to generate high impact graphics and movies.

2019.2 CSD Release (May 2019)

Data Updates:

  • 14,702 new entries
  • 14,140 new unique structure determinations
  • Enables access to over one million entries through ConQuest searches, Mercury analyses and other knowledge-based research (note: that this release did not allow access to one million unique crystallographic structures).
  • Read more about this data release.


  • Support in the CSD Python API for all types of 3D query available in ConQuest.
  • Extension of the CSD-CrossMiner database to include nucleic acid-containing structures from the PDB. Searching these structures was augmented by the addition of new, nucleotide specific feature definitions.
  • Improvements enabling searching of the Protein Data Bank (PDB). This release provided a set of augmentations to the CSD Python API that make the software better at handling proteins. These changes mean that the same 3D queries can be constructed and run against individual protein structures.

2019.1 CSD Release (February 2019)

Data Updates:

  • Size of the CSD reaches over 977,000 unique structure determinations
  • 20,808 new entries
  • Read more about this data release.


  • The launch of polyhedral display in Mercury, facilitating access to a simple, intuitive option for users to visualise metal-organic structures using polyhedral representations. The inclusion of a polyhedral display style, partnered with Mercury’s structure exploration features and high-resolution image output using POV-Ray, make generating high-quality graphics for metal-organic structures effortless.
  • The first release of the CSD Python API that is Python 3 compatible. Additional Python 3.7 packages are now available to download and install into your preferred Python environment.